Magnesium transport systems: genetics and protein structure (a review).

Magnesium is unique among biological cations. Its volume change from hydrated cation to atomic ion is over an order of magnitude larger than that of any other biological cation. This volume change presents particular problems for a magnesium transport system and suggests that these systems may be significantly different from other classes of ion transport systems. Detailed study of Mg2+ transport in complex organisms is limited by severe technical problems. However, molecular genetic techniques have enabled the isolation of three Mg2+ transport systems from the Gram-negative bacterium Salmonella typhimurium. The MgtA and MgtB transport systems are members of the P-type ATPase superfamily of transporters but possess unique characteristics among members of this family. The CorA transport protein is the first member of an entirely new class of transport proteins. In addition, another completely new family of Mg2+ transport proteins have been identified that is present in both Gram-negative and Gram-positive bacteria. Characterization of these transporters should provide substantial insight into Mg2+ transport and cellular Mg2+ homeostasis.

Role of the TonB amino terminus in energy transduction between membranes.

Escherichia coli TonB protein is an energy transducer, coupling cytoplasmic membrane energy to active transport of vitamin B12 and iron-siderophores across the outer membrane. TonB is anchored in the cytoplasmic membrane by its hydrophobic amino terminus, with the remainder occupying the periplasmic space. In this report we establish several functions for the hydrophobic amino terminus of TonB. A G-26-->D substitution in the amino terminus prevents export of TonB, suggesting that the amino terminus contains an export signal for proper localization of TonB within the cell envelope. Substitution of the first membrane-spanning domain of the cytoplasmic membrane protein TetA for the TonB amino terminus eliminates TonB activity without altering TonB export, suggesting that the amino terminus contains sequence-specific information. Detectable TonB cross-linking to ExbB is also prevented, suggesting that the two proteins interact primarily through their transmembrane domains. In vivo cleavage of the amino terminus of TonB carrying an engineered leader peptidase cleavage site eliminates (i) TonB activity, (ii) detectable interaction with a membrane fraction having a density intermediate to those of the cytoplasmic and outer membranes, and (iii) cross-linking to ExbB. In contrast, the amino terminus is not required for cross-linking to other proteins with which TonB can form complexes, including FepA. Additionally, although the amino terminus clearly is a membrane anchor, it is not the only means by which TonB associates with the cytoplasmic membrane. TonB lacking its amino-terminal membrane anchor still remains largely associated with the cytoplasmic membrane.

Analysis of Escherichia coli TonB membrane topology by use of PhoA fusions.

Alkaline phosphatase (PhoA) fusions to TonB amino acids 32, 60, 125, 207, and 239 (the carboxy terminus) all showed high PhoA activity; a PhoA fusion to TonB amino acid 12 was inactive. The full-length TonB-PhoA fusion protein was associated with the cytoplasmic membrane and retained partial TonB function. These results support a model in which TonB is anchored in the cytoplasmic membrane by its hydrophobic amino terminus, with the remainder of the protein, including its hydrophobic carboxy terminus, extending into the periplasm.

A mutation in the amino terminus of a hybrid TrpC-TonB protein relieves overproduction lethality and results in cytoplasmic accumulation.

We have developed a selection for mutations in a trpC-tonB gene fusion that takes advantage of the properties of the plasmid-encoded TrpC-TonB hybrid protein. The TrpC-TonB hybrid protein consists of amino acids 1 through 25 of the normally cytoplasmic protein, TrpC, fused to amino acids 12 through 239 of TonB. It is expressed from the trp promoter and is regulated by the trpR gene and the presence or absence of tryptophan. Under repressing conditions in the presence of tryptophan, the trpC-tonB gene can restore phi 80 sensitivity to a tonB deletion mutant, which indicates that TrpC-TonB can be exported and is functional. High-level expression of TrpC-TonB protein in the absence of tryptophan results in virtually immediate cessation of growth for strains carrying the trpC-tonB plasmid. By selecting for survivors of the induced growth inhibition (overproduction lethality), we have isolated a variety of mutations. Many of the mutations decrease expression of the TrpC-TonB protein, as expected. In addition, three independently isolated mutants expressing normal levels of TrpC-TonB protein result in a Gly----Asp substitution within the hydrophobic amino terminus of TonB. The mutant proteins are designated TrpC-TonBG26D. The mutations are suppressed by prlA alleles, known to suppress export (signal sequence) mutations. TrpC-TonB proteins carrying the Gly----Asp substitution accumulate in the cytoplasm. We conclude that the Gly----Asp substitution is an export mutation. TrpC-TonBG26D protein has been purified and used to raise polyclonal antibodies that specifically recognize both TrpC-TonB protein and wild-type TonB protein.

Effect of lasalocid, monensin or thiopeptin on lactic acidosis in cattle.

Lasalocid, monensin or thiopeptin was administered intraruminally each at .33, .65 or 1.3 mg/kg body weight and evaluated for its effectiveness in preventing experimentally induced lactic acidosis in cattle. Four rumen-fistulated cattle were used for each dosage level and the design was a 4 x 4 Latin square with each animal receiving lasalocid, monensin, thiopeptin or no antibiotic. Acidosis was induced by intraruminal administration of glucose (12.5 g/kg body weight). Control cattle exhibited the typical drop in rumen pH and concurrent increases in L(+) and D(-) lactate concentrations commonly observed in cases of lactic acidosis. Alkali reserves were depleted in the control cattle as evidenced by a decrease in blood bicarbonate and a negative shift in base excess. In all three trials, cattle given lasalocid had higher rumen pH and lower lactate concentrations than did control cattle or cattle given monensin or thiopeptin. Cattle given monensin had a significantly higher rumen pH and a lower lactate concentration than the controls only at the .65 and 1.3 mg/kg body weight dosages, whereas thiopeptin was effective only at the 1.3-mg dosage. Concentrations of total VFA in rumen fluid decreased in the controls but remained unchanged in cattle given antibiotics. A significant reduction in the molar proportion of acetate and an increase in the molar proportion of propionate were observed in the rumen fluid of the cattle given antibiotics. Colony counts of Streptococcus bovis and Lactobacillus were significantly reduced in rumen fluid of cattle given 1.3 mg antibiotic/kg body weight. Counts of lactate-utilizing bacteria increased in both control cattle and cattle given antibiotics. Cattle given antibiotics showed no evidence of lacticacidemia, hemoconcentration or change in acid-base balance.